Sampling Techniques and Sample Size

Determination

1
Objectives

— Learn the reasons for sampling

— Develop an understanding about different sampling
methods
— Distinguish between probability & non probability
sampling
— Discuss the relative advantages & disadvantages of each
sampling methods
— list the factors influencing the sample size
— calculate the sample size using appropriate formulae

2
 Sampling

 A sample is “a smaller (but hopefully representative)

collection of units from a population used to
determine truths about that population” .

 Why sample?

3
 Why sample?
 Cost in terms of money, time and manpower
 Accessibility
 Utility e.g. to do diagnostic laboratory test you
don’t draw the whole of patient’s blood.
A census is a sample consisting of the entire population.
Even though a census is not full proof, it gives detailed
information about every small area of the population.
It has the following disadvantages:
 Expensive
 Takes a long time
 Cumbersome & therefore inaccurately done ( a careful sample
produces a more accurate data than a census.)

4
 Sampling…..
 Sampling is the process of selecting a representative sample
from populations.
 It Selecting cases (elements)—or locating people (or other units of analysis)
—from a target population in order to study the population.

sampling
Sample
Inference
Population
5
Cont’d
 The process of obtaining information from a subset (sample) of a larger
group (population)
 The results for the sample are then used to make estimates of the larger
group
 Faster and cheaper than asking the entire population
 Two keys
1. Selecting the right people
 Have to be selected scientifically so that they are representative of the population
2. Selecting the right number of the right people
 To minimize sampling errors I.e. choosing the wrong people by chance

6
Population Vs. Sample
Population of Interest

Population Sample

Sample
Parameter Statistic

We measure the sample using statistics in order to draw

inferences about the population and its parameters.

7
Characteristics of Good Samples
o Representation
 Sample surveys are almost never conducted for the
purposes of describing the particular sample under
study. Rather they are conducted for purposes of
understanding the larger population from which the
sample was initially selected
 A great deal of work has been done over the years in
developing sampling methods that provide
representative samples for the general population.
E.g. international survey programs such as the DHS series, EPI
coverage surveys have perfected the art of household sampling.
8
Characteristics of Good Samples

 3 factors that influence sample representativeness

 Sampling procedure
 Sample size
 Participation (response)

 When might you sample the entire population?

 When your population is very small
 When you have extensive resources
 When you don’t expect a very high response
o Accessible
o Low cost
9
 Basic Terms

 population (also called source population or target

population): is a group of individuals persons, objects, or
items from which samples are taken for measurement.

 It refers to the entire group of individuals or objects to

which researchers are interested in generalizing the
conclusions.

10
Basic term cont’d….

11
 Basic Terms cont’d…

 Census: Obtained by collecting information about each

member of a population. Studying the whole population and
requires a great deals of time, money and energy.

 Sample survey: study sample and draw conclusions about

populations. It is cheaper in terms of cost, practical &
convenient in terms of technicalities, saves time & energy.

12
 Basic Terms cont’d…

 Sampling Frame: is the list of people from which the

sample is taken. It is the list from which the potential
respondents are drawn.

It should be comprehensive, complete and up-to-date.

Examples of sampling frame: Electoral Register;
Postcode Address File; telephone book and so on.
 Probability samples: With probability sampling methods, each
population element has a known (non-zero) chance of being
chosen for the sample.

13
 Basic term cont’d….

 Non-probability samples: With non-probability sampling

methods, we do not know the probability that each
population element will be chosen, and/or we cannot be sure
that each population element has a non-zero chance of being
chosen
 Sampling unit - the unit of selection in the sampling process
 Study unit (study subjects)- the unit on which information
is collected or on which observations are made. E.g. Familiar
examples are families, towns, litters, branches of a company,
individual subjects or schools.
14
 Basic term cont’d….

• The sampling unit is not necessarily the same as the study

unit.
• If the objective is to determine the availability of latrine,
• then the study unit would be the household;
• If the objective is to determine the prevalence of
trachoma, then the study unit would be the individual.
Sampling fraction (Sampling interval) - the ratio of the
number of units in the sample to the number of units in the
reference population (N/n)

15
Hierarchy of sampling
Study subjects
The actual
participants in
the study
Sample
Subjects who are
selected

Sampling Frame
The list of potential subjects
from which the sample is
drawn

Source population
The Population from whom the study
subjects would be obtained

Target population 16

The population to whom the results would be applied

 Errors in statistical Study

A sample is expected to mirror the population from which it

comes, however, there is no guarantee that any sample will be
precisely representative of the population.
No sample is the exact mirror image of the population .

Sampling or Random

Errors
Non-sampling or
systematic
17
1. Sampling error

– random error- the sample selected is not

representative of the population due to chance
– The uncertainty associated with an estimate that is based
on data gathered from a sample of the population rather
than the full population is known as sampling error.
– Sampling errors are the random variations in the sample
estimates around the true population parameters.

18
Sampling error cont’d…

the level of it is controlled by sample size

a larger sample size leads to a smaller sampling error. it
decreases with the increase in the size of the sample,
and it happens to be of a smaller magnitude in case of
homogeneous population.
When n = N ⇒ sampling error = 0

 Can not be avoided or totally eliminated

19
 Sampling error cont’d…

why do sample estimates have uncertainty associated

with them? There are two reasons.
 Estimates of characteristics from the sample data can
differ from those that would be obtained if the entire
population were surveyed.

 Estimates from one subset or sample of the population

can differ from those based on a different sample from
the same population (sample to sample variations).

20
 The cause of sampling error

Chance: main cause of sampling error and is the error that

occurs just because of bad luck.

Sampling bias: Sampling bias is a tendency to favor the

selection of participants that have particular characteristics.
The chance component (sometimes called random error)
exists no matter how carefully the selection procedures are
implemented, and the only way to minimize chance-
sampling errors is to select a s u ff i c i e n t l y l a rg e
sample.
21
 2. Non Sampling Error

It is a type of systematic error in the design or conduct of a

sampling procedure which results in distortion of the sample, so
that it is no longer representative of the reference population.

We can eliminate or reduce the non-sampling error (bias) by

careful design of the sampling procedure and not by increasing
the sample size.

It can occur whether the total study population or a sample is

being used.

22
 Non-sampling Error……
o The basic types of non-sampling error
 Non-response error
 Response or data error
o A non-response error occurs when units selected as part of the
sampling procedure do not respond in whole or in part
 If non-respondents are not different from those that did
respond, there is no non-response error
 When non-respondents constitute a significant proportion of
the sample (about 15% or more

23
Non-sampling Error…….
o A response or data error is any systematic bias
that occurs during data collection, analysis or
interpretation
 Respondent error (e.g., lying, forgetting, etc.)
 Interviewer bias
 Recording errors
 Poorly designed questionnaires

24
Non-Sampling Error cont’d …

Systematic error makes survey results unrepresentative of the

target population by distorting the survey estimates in one
direction.

Random error can distort the results in any given direction but
tend to balance out on average
Thus, the total survey error

sampling error + non-sampling error

25
 Advantage of sampling

We obtain a sample rather than a complete enumeration (a

census ) of the population for many reasons.
 Feasibility it may be the only feasible method of
collecting data
 Reduced cost sampling reduces demands on resource
such as finance, personal and material
 Greater accuracy sampling may lead to better accuracy
of collecting data.
 Greater speed data can be collected and summarized
more quickly

26
 Disadvantage of Sampling

If sampling is biased, or not representative or too small the

conclusion may not be valid and reliable

If the population is very large and there are many sections and
subsections, the sampling procedure becomes very complicated

If the researcher does not possess the necessary skill and

technical knowledge in sampling procedure, then the outcome
will be devastated.

27
 Characteristics Of A Good Sample Design

From what has been stated above, we can list down the characteristics
of a good sample design as:
Sample design must result in a truly representative sample.
Sample design must be such which results in a small sampling
error.
Sample design must be viable in the context of funds available for
the research study.
Sample design must be such so that systematic bias can be
controlled in a better way.
Sample should be such that the results of the sample study can be
applied, in general, for the universe with a reasonable level of
confidence.
28
 Types of Sampling

How we Selecting the right subjects

o The sample that we draw for our study

determines the generalizability of our
findings.
o Sample should to have a good representation
of the population.

29
 Types of Sampling Methods

Sampling Method

Non-Probability Probability Samples

Samples
Simple Stratified
Random
Quota
Judgemental
Systematic Cluster

Convenience
Multistage Random
Sampling
30
 Probability Sampling Method …

The random ("equal chance“) and "independent" components of random

sampling are what makes us confident that the sample has a reasonable
chance of representing the population

What does it mean to be independent? The researchers select each

person for the study separately.

Let us say you were asked to participate in an experiment, enjoyed it,

and told your friends to contact the researcher to volunteer for the study.

 This would be an example of non-independent sampling.

31
 Probability Sampling Method cont’d …

In probability sampling
A sampling frame exists or can be compiled.
should have an equal or at least a known or nonzero chance
of being included in the sample.
Generalization is possible (from sample to population)
 Simple Random Sampling,
 Systematic Sampling,
 Stratified Random Sampling,
 Cluster Sampling
 Multistage Sampling.
32
 1. Simple Random Sampling(SRS)

Simple random sampling is the most straightforward of the

random sampling strategies.

To use SRS there should be

o sampling frame for the population
1
o All possible samples of “n” subjects are equally likely ( n ) to occur.
o population is small, relatively homogeneous & readily available

33
 Simple Random Sampling cont’d …

Procedures to select the sample

 The specific procedures that you follow may vary depending
on your resources, but all involve some type of random
process. Depending on the complexity of the population, we
can use different tools to select “n” samples from the given
sampling frame.
 These are lottery method,
 table of random number (they are available in the appendix
of many research methods and statistics textbooks) or
 computer generated random number.

34
 Simple Random Sampling cont’d …
Lottery method is appropriate if the total population is not too
large, otherwise if the population is too large then it will be very
difficult to use lottery method.
Thus, table of random number or computer generated random
number is the feasible method to be used.
Sampling schemes may be
o without replacement- no element can be selected more than once in the
same sample,  N possible samples.

n 

 
o with replacement- an element may appear multiple times in the one sample
possible samples.n
N
35
 Example

Assume that the total number of patients who visit Gondar

University Hospital for the last six months is “N”. We want to
see the prevalence of TB among those patients who visited the
hospital.

ti on is
pop ul a
ou g h t h e
m a y not be
Th
g e n e o u s, there
homo f r ame .
avail a bl e
so
what?

36
 2. Systematic Random Sampling

Systematic sampling is thought as random, as long as the periodic interval is

determined beforehand and the starting point is random
A method of selecting sample members from a larger population according to
a random starting point and a fixed, periodic interval.
Typically, every nth member is selected from the total population for inclusion
in the sample population.
It is frequently chosen by researchers for its simplicity and its periodic
quality.
it needs the population to be homogeneous, however the method does not
require frame.

37
Steps in systematic sampling:

Define the population

Determine the desired sample size (n)
List the population from 1 to N
Determine K, where k=N/n
Select a random number between 1 and k, let us denote this number by “a”
Starting at a, take every Kth number on the list until the desired sample is
obtained.
Then the selected list will be
a, a+k, a+2k, a+3k, …, a+(n-1)k

Note: Systematic sampling should not used when a cyclic repetition is

inherent in the sampling frame

38
E.g. systematic sampling

• N = 1200, and n = 60
sampling fraction = 1200/60 = 20
• List persons from 1 to 1200
• Randomly select a number between 1 and 20
(e.g. 8)
• 1st person selected = the 8th on the list
• 2nd person = 8 + 20 = 28th list e.t.c.

39
 Systematic sampling….

o It relies on arranging the target population according to some

ordering scheme and then selecting elements at regular
intervals through that ordered list.

o Systematic sampling involves a random start and then

proceeds with the selection of every kth element from then
onwards. In this case, k =(population size/sample size).
o It is important that the starting point is not automatically the
first in the list, but is instead randomly chosen from within
the first to the kth element in the list.

40
 Though the frame available, the population may
not be homogeneous, so what?

41
 3. Stratified Random Sampling

Stratified random sampling is used when we have subgroups in

our population that are likely to differ substantially in their
responses or behavior (i.e. if the population is heterogeneous).

In stratified random sampling, the population is first divided into

a number of parts or 'strata' according to some characteristic,
chosen to be related to the major variables being studied.

For example, you are interested in visual-spatial reasoning and

previous research suggests that men and women will perform
differently on these types of task
42
 Stratified Random Sampling cont’d…

So, you divide your sample into male and female members and
randomly select the required sample size within each subgroup
(or "stratum")

With this technique, you are guaranteed to have enough of each

subgroup for meaningful analysis.

Often we used simple random sampling to select a sample from

each strata after stratification.

43
Steps involve in stratified sampling method:
Define the population
Determine the desired sample size
Identify the variable and subgroups (strata) for which you want to
guarantee appropriate representation (either proportional or equal)

Classify all members of the population as a member of one of the

identified subgroups
Randomly select (using simple random sampling or others) an appropriate
number of individuals from each subgroup.

Then the total sample size will be the sum of all samples from each
subgroup. 44
 There are two methods to get the study subject from each subgroup,
proportional allocation or
equal allocation.
We use proportional allocation technique when our subgroups vary dramatically in size
in our population
 Let N be total population and N1, N2 . . . . Nk be the subtotal population for strata 1, 2,
…. K respectively. Moreover let n be the total sample size and n1, n2…..nk be th
subsample for strata 1, 2…..k respectively in which N = N1 + N2 +….. …+ NK
and n = n1 + n2 + …………..+ nk
Then the subsample “ni “which will be selected from subgroup Ni can be computed by

n  Ni
ni  where i  1, 2, 3........k
N
45
The higher the population in the subgroup, the higher the
sample size will be.

However, equal allocation will be used if the total population

from each subgroup is approximately equal.

46
Advantage of stratified sampling over simple random sampling

The representativeness of the sample is improved. That is,

adequate representation of minority subgroups of interest can
be ensured by stratification and by varying the sampling
fraction between strata as required.

DEMERIT
Sampling frame for the entire population has to be prepared
separately for each stratum.

47
 4. Cluster Random Sampling

In this sampling scheme, selection of the required sample is done on groups

of study units (clusters) instead of each study unit individually.
The sampling unit is a cluster, and the sampling frame is a list of these
clusters.
If the study covers wide geographical area, using the other methods will be
too costly.
The idea is, divided the total population in to different clusters and then the
unit of selection will be cluster.
Therefore, total population in the selected cluster will be taken as the sample.

48
 Steps in cluster sampling are:

 Define the population

 Determine the desired sample size
 Identify and define a logical cluster (can be kebele, Got, residence,
and so on)
 Make a list of all clusters in the population
 Estimate the average number of population number per cluster
 Determine the number of clusters needed by dividing the sample size
by the estimated size of the cluster
 Randomly select the required number of clusters (using table of
random number as the total number of clusters is manageable)
 Include in the sample all population in the selected cluster.
49
Consider the following graphical display:

50
 5. Multistage Random Sampling

This is the most complex sampling strategy.

The researcher combines simpler sampling methods to address sampling
needs in the most effective way of possible.
Example 1,
 The administrator might begin with a cluster sample of all schools in the
district.
 Then he might set up a stratified sampling process within clusters.
 Within schools, the administrator could conduct a simple random sample
of classes or grades.
 By combining various methods, researchers achieve a rich variety of
results useful in different contexts.

51
 Non-Probability Sampling Method

In the presence of constraints to use probability sampling

strategies, the alternative sampling method is non-probability
sampling method.

Non-probability sampling strategies are used when it is

practically impossible to use probability sampling strategies.

Non-probability sampling is sampling procedure which does not

afford any basis for estimating the probability that each item in
the population has of being included in the sample.

52
Cont’d……….

 Subjective units of population have a zero or unknown

probability of selection before drawing the as sample. Hence
obtained a non-representative samples.
 Sampling error can not be computed
 Survey results cannot be projected to the population

Advantages
 Cheaper and faster than probability
 Reasonably representative if collected in a thorough manner
53
1. Judgment Sampling/ Purposive sampling

The researcher selects the sample based on judgment. A researcher

exerts some effort in selecting a sample that seems to be
most appropriate for the study.

This is used primarily when there is a limited number of

people that have expertise in the area being researched

54
2. Convenience Sampling

Convenience sampling selects a particular group of people but

it does not come close to sampling all of a population.

The sample would generalize only to similar programs in

similar cities.
It looks just like cluster sampling.
The major difference is that the clusters of research
participants are selected by convenience rather than by a
random process.

55
Cont’d………..

 Sometimes known as grab or opportunity sampling or

accidental or haphazard sampling.

 A type of non probability sampling which involves the sample

being drawn from that part of the population which is close to
hand. That is, readily available and convenient.

 The researcher using such a sample cannot scientifically make

generalizations about the total population from this sample
because it would not be representative enough.

 This type of sampling is most useful for pilot testing.

56
3. Quota sampling

It is a method that ensures a certain number of sample units

from different categories with specific characteristics are
represented. The investigator interviews as many people in
each category of study unit as he can find until he has filled his
quota.
It is the non-probability equivalent of stratified sampling. This
differs from stratified sampling, where the stratums are filled
by random sampling.
The population is first segmented into mutually exclusive sub-
groups, just as in stratified sampling.

57
Cont’d

 Then judgment used to select subjects or units from

each segment based on a specified proportion.

 For example, an interviewer may be told to sample

200 females and 300 males between the age of 45 and
60.

 It is this second step which makes the technique one

of non-probability sampling.

58
Cont’d
 In quota sampling the selection of the sample is non-random.

 For example interviewers might be tempted to interview

those who look most helpful. The problem is that these
samples may be biased because not everyone gets a chance of
selection.

 This random element is its greatest weakness and quota

versus probability has been a matter of controversy for many
years

59
 4. Snowball sampling

It is a special non-probability method used when the

desired sample characteristic is rare.
Snowball sampling relies on referrals from initial
subjects to generate additional subjects.
What we need to do in case of snowball sampling is
that first identify someone who meets the criteria and
then let him/her bring the other he/she knew.

60
Cont’d
 While this technique can dramatically lower search costs,
it comes at the expense of introducing bias because the
technique itself reduces the likelihood that the sample will
represent a good cross section from the population.

 Selection of additional respondents is based on

referrals from the initial respondents.
 friends of friends
 Used to sample from low incidence or rare populations.

61
 Sample Size Determination

Determining the sample size for a study is a crucial component

of study to include sufficient numbers of subjects so that
statistically significant results can be detected.

"How large a sample do I need?“

The answer will depend on the aims, nature and scope of the
study and on the expected result. All of which should be
carefully considered at the planning stage.

62
Sample……

o If sample (“n”) is
 Large
 Increase accuracy
 Costy / complex
Take
Optimum
 Small sample
o Decrease accuracy
o Less costy

How ?

63
Factors to determine sample size
 Size of population
 Resources – subjects, financial, manpower
 Method of Sampling- random, stratified
 Degree of difference to be detected
 Variability (S.D.) – pilot study, historical
 Degree of Accuracy (or errors)
- Type I error (alpha) p<0.05
- Type II error (beta) less than 0.2 (20%)
- Power of the test : more than 0.8 (80%)
 Statistical Formulae
 Dropout rate, non-compliance to Rx

64
o Sample size determination depending on outcome variables.

There are three possible categories of outcome variables.

 The first is where the variable of interest has only two
alternatives response: yes/no, dead/alive, vaccinated/not
vaccinated and so on.
 The second category covers those outcome variable with
multiple, mutually exclusive alternatives responses, such as
marital status, religion, blood group and so on.
 For these two categories of outcome variables, the data are
generally express as percentages or rates.
 So we can use percentage to compute the sample size.

65
 The third category covers continuous response variables
such as birth weight, age at first marriage, blood
pressure and cerium uric acid level, for which
numerical measurement are usually made.

 In this case the data are summarize in the form of means

and standard deviations or their derivatives.

66
 Sample Size………...

There are several approaches to determining the sample size.

Depending on the type of response variable, whether it is

categorical or continuous, we will have two sets of formulas.

The sample size determination formulas come from the formulas

for the maximum error of the estimates and is derived by solving
for n.

67
Sample for Single population
To estimate sample size for single survey using simple
or systematic random sampling, need to know:
oEstimate of the prevalence of the outcome
o Precision desired
o Design effect
o Size of total population
oLevel of confidence (always use 95%)

68
 Sample size for single population mean

This is the condition in which the research question is about

mean.
Standard deviation () of the population: It is rare that a
researcher knows the exact standard deviation of the population.
Typically, the standard deviation of the population is estimated:
 from the results of a previous survey,
 from a pilot study,
 from secondary data,
 from judgment of the researcher.

69
Maximum acceptable difference (w): This is the maximum
amount of error that you are willing to accept.
Desired confidence level (Z/2 ) : is your level of certainty that
the sample mean does not differ from the true population mean
by more than the maximum acceptable difference. Commonly
we use a 95% confidence level.
Then the sample size determination formula for single
population mean is defined by:
z22   2
n 
w2
70
Sample size for single population mean cont’d…

 Where
 α= The level of significance which can be obtain as 1-
confidence level.
 σ=Standard deviation of the population
 w= Maximum acceptable difference
 z α/2 = The value under standard normal table for the
given value of confidence level

71
Sample Size for Single Population Proportion

This is the situation in which the variable of interest is categorical.

Three questions must be answered to determine the sample size for
single population proportion:
Best estimate of population proportion of the variable of interest :
Make your best estimate of what the actual percent of the survey
characteristic is.
The possible source of this proportion are:
 from the results of a previous study,
 item from a pilot study,
 item judgment of the researcher.
 item Simply taking 50%

72
Then the formula for the sample size of single population proportion is
defined as:

z22 * p (1  p )
n 2
w

Where α = the level of significance which can be obtained as 1- confidence level.

P = best estimate of population proportions
W = maximum acceptable difference
z 2 the value under standard normal table for the given value of confidence
level

73
 Example 1

One of MPH student want to conduct a research on the prevalence of ANC utilization
of mothers in a district. Given that the prevalence from the previous study found to be
45.7% , what will be the sample size he should take to address his objective?
Solution:
 Margin of error d= 5%
 A confidence level of 95% will give the value of as Zα/2=1.96.
 Then using the formula :

2 2

 Z  P (1  P ) 
 Z 0.05  0.457(1  0.457)
n  2 
  2 

W2 0.05 2


1.96  0.457(0.543)
2

0.05 2
 382

74
 Some Considerations
 The final sample size will be corrected for
 Nonresponse, lost to follow up, lack of compliance and so on
 Consider the total size of the population (N): if N <10000 then we
need correction the formula which is defined by

no
nf 
n
1 o
N

 Where nf = final sample size, no = sample size from the above

formula and N total population.
 Take the design effect in to account if needed

75
Incorrect sample size will lead to
o Wrong conclusions
o Poor quality research (Errors)
o Type II error can be minimized by increasing the sample size
o Waste of resources
o Loss of money
o Ethical problems
o Delay in completion

76
 Example 2

Midwifery graduate student wants to do her thesis work on the

title “assessment of the outcome of pregnancy among women
who visited a hospital gynecology and obstetrics ward for the
year 2010”
What will be the sample size she should take for this study?

77
78